Skip to content
Restaurant · Booking flow · 2026

Verre

A restaurant reservation flow rebuilt around one constraint: reduce the number of taps between 'I want a table' and a confirmed booking, without losing any of the covers-management logic a real host stand needs.

Client
Self-initiated — a restaurant reservation concept
Role
Solo — design and build
Duration
One week (Week 50), full-time evenings
Next.jsTailwind CSS

The brief

The Dribbble reference was a beautiful multi-step reservation wizard — five screens deep. Real usage data from comparable restaurant sites shows drop-off compounds badly past two steps on mobile, so the brief was to compress the flow without losing party-size, date, and seating-preference capture.

The build

  1. 01

    Collapsed the five-step wizard into a single scrollable form with progressive disclosure — later fields appear only once earlier ones are valid, rather than gating behind a 'Next' button.

  2. 02

    Built the date/time picker as a horizontal-scroll strip of available slots instead of a calendar modal, since most bookings are 'tonight' or 'this weekend,' not far-future planning.

  3. 03

    Kept the confirmation state inline (no redirect to a separate confirmation page) so the booking felt instant.

What didn’t work

An initial version auto-selected the 'most popular' time slot to reduce decisions — user testing read this as pushy, not helpful, since it felt like the studio choosing for the diner. Removed the auto-selection; the slot strip now shows availability with zero default bias.

Related Templates

Related Builds

  • Verre — restaurant reservation flow